Why does intensive insulin treatment reduce the incidence of microvascular complications but not the frequency or severity of atherosclerosis in IDDM? We believe that this paradox Is In part attributable to conventional subcutaneous Insulin therapy's (CIT) atherogenic effects on lipoprotein transport and remodeling. This unphysiologic mode of Insulin delivery leads to peripheral hyperinsulinism, stimulation of lipoprotein lipase (LpL), and alteration of the surface properties of larger VLDL particles, Increasing their affinity for cholesterol ester protein (CETP). This Interaction augments the exchange of neutral lipids between VLDL and HDL. To support this hypothesis, we have found that in well controlled CIT-treated IDDM patients: 1) basal LpL activity Is Increased; 2) the transfer of cholesteryl ester (CET) from HDL to VLDL estimated both by Isotopic and mass transfer assays is accelerated; 3) the core lipid composition of VLDL and HDL Is altered In a manner consistent with Increased CET taking place in vivo; 4) this acceleration in CET results from the abnormal behavior of the VLDL-1 subfraction; 5) this disturbance in CET In 5 patients was reversed by when Insulin treatment was delivered by an implanted intraperitoneal pump (IP). We have the extraordinary opportunity to test our hypothesis that the potentially atherogene changes in lipoprotein composition and function we have found are secondary to iatrogenic hyperinsulinism by studying the same well-controlled IDDM patients first after intensive CIT and again serially over a period of one year following IP treatment which by delivering insulin preferentially to the portal system reduces peripheral hyperinsulinism. The data obtained In these IP-treated patients will be compared to those obtained from matched IDDM CIT-treated patients with the same optimal level of glycemic control and nondiabetic controls. We Intend to demonstrate the differing effects of IP and CIT on; 1) CET, a key step in reverse cholesterol transport; 2) other systems In plasma (LpL, hepatic lipase, and LCAT which normally remodel lipoproteins but if pathologically active can disturb their composition and function and generate increased amounts of certain molecular species of cholesteryl ester and lysolecithin; 3) the capacity of IDDM VLDL subfractions to deliver cholesterol to THP-1 cells which resemble human monocyte-derived macrophages. We seek to determine whether the more physiologic route of IP Insulin administration can correct the spectrum of disturbances we have found earlier In the function and composition of VLDL, and other IDDM lipoproteins which our studies suggest make a significant contribution to accelerated atherogenesis in IDDM.